In this masterclass, the expert speakers will discuss how even innocent-seeming "irony" and unintentional spread of hateful content can act like a contagious pathogen online and offline if not detected.

Identifying high-risk content of various types that contribute to hate speech, hate crime, and violent extremism can severely reduce harms to online and offline communities, while preventing the normalization of highly problematic conduct and content.

Taking basic steps such as updating tools for detection and consumer education on what good behavior online means, platforms can build a positive and safe space that limits toxic content.

Key takeaways:

• Tactics, tools and metrics platforms can use to encourage positive online behavior
• How to identify toxic behavior before it escalates
• The reasons hate speech is difficult to detect on platforms and what you can do to detect it
• Examples of encouraging positive online behavior and why it's important for platforms to do this
• Why data labeling, detection of complex behaviors, user-level and real-time actioning, analytics are so important to platforms